- From: Georgi Kobilarov <georgi.kobilarov@gmx.de>
- Date: Fri, 29 Jan 2010 14:10:30 +0100
- To: "'Kingsley Idehen'" <kidehen@openlinksw.com>
- Cc: <public-lod@w3.org>, <semantic-web@w3.org>
Hi Kingsley, > I've lookup a URI using your service: > http://uberblic.org/resource/a8c82706-e855-490c-96eb-6061b6d0c3c8#thing > > I can't seem to find an graph level relationships with its original sources > (DBpedia, BBC etc.) . Is this me, your solution design and implementation > choices, or something else? There are http://uberblic.org/meta/source_uri links to the original sources in .rdf and .trig output: http://uberblic.org/resource/a8c82706-e855-490c-96eb-6061b6d0c3c8.rdf If the original resource was a RDF resource (e.g. DBpedia, Geonames, BBC), there are owl:sameAs links as well. E.g. London with a link to Geonames: http://uberblic.org/resource/e4c9b34a-9680-4a3d-a922-7aab04b1532b.rdf If you follow the "Source Details" link at the top right hand side in HTML pages, you'll see our triples grouped by source document. > At the very least, I would expect to be able to navigate from your linked data > space to other linked data spaces, starting with your URIs. > > Also, will you be using <link/> in our HTML pages (re. auto-discovery > patterns) or is it strictly content negotiation? Ditto RDFa within your HTML > pages etc.. Yes, thanks for the suggestion. We'll add <link/> tags to HTML pages in a bit... We are probably not going to provide RDFa in the near future, since I believe it would just increase the content-length of HTML pages, but not provide any added benefit since RDF output is available separately. In large resources (like London above), we sample resource links in the HTML output (with links to "and X more..."), and I'm not sure how RDFa would fit in there. Best, Georgi -- Georgi Kobilarov Uberblic Labs Berlin www.georgikobilarov.com
Received on Friday, 29 January 2010 13:11:02 UTC